MIT OpenCourseWare
http://ocw.mit.edu
3.23 Electrical, Optical, and Magnetic Properties of Materials
Fall 2007
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.
Homework # 2
Nicolas Poilvert & Nicola Marzari
September 19, 2007
Homework is due on Wednesday September 26th, 5pm
1 The power of spinach
MIT researchers have been able to use the Photosystem 1 in spinach chloroplasts
to convert sunlight to energy into a solid-state electronic spinach sandwich de­
vice that may one day power laptops and cell phones. What would be the size
of the confining region for the electron responsible for the green color of spinach
? (Hint: To study this problem, assume that the electronic transition is from
the ground state to the first excited state. You can think at Photosystem 1 as
being a 1-d infinite well)
2 Molecular orbitals and the Linear Combinaition
of Atomic Orbitals method (LCAO)
Background on LCAO In this problem, we will be interested in character­
izing the two lowest lying molecular orbitals of an electron in the Hydrogen
molecule H2 . To find those molecular orbitals, we will use a well-known tech­
nique in Quantum Chemistry called the Linear Combinaition of Atomic Orbitals
(LCAO). The spirit of the technique is the following: Imagine that you take two
hydrogen atoms far from each other. In that case each atom has one elec­
tron, and the wavefunction
of each electron is described by a 1s atomic orbital,
r
φ1s = √ 1 3 e− a0 , where a0 is the Bohr radius. When one then reduces the
πa0
distance between the two hydrogen atoms up to atomic dimensions, the ground
state wavefunction of each electron is no longer described by a 1s orbital because
an electron localized around one of the proton can hop to the second proton of
the H2 molecule. So what is the wavefunction then? The LCAO technique ap­
plied to the case of the H2 molecule consists in writing the total wavefunction of
an electron as a linear combinaison of two 1s orbitals, one centered around each
proton. The new wavefunction is then called a molecular orbital because it
spreads over the entire molecule and is not localized around a particular atom.
Hence the molecular orbital looks like this:
ψ = c1 φ1s (1) + c2 φ1s (2)
1
where (1) and (2) stand for proton 1 and 2. We can compactly represent this
linear combinaison as a 2-dimensional vector in the basis of φ1s (1) and φ1s (2)
as:
�
�
c1
ψ=
c2
Expressing the Hamiltonian of an electron Let’s call Ĥ the hamiltonian
of a single electron in an H2 molecule. Since we expressed the electronic wavefunction as a linear combinaition of φ1s (1) and φ1s (2) only, we need to find
the hamiltonian matrix in the basis formed by those two atomic orbitals if we
want to be able to find the eigenenergies (i.e the molecular energies) and the
corresponding eigenfunctions (i.e the molecular orbitals). So we see that instead
of expressing the hamiltonian in an infinite and complete basis, we chose a
two-dimensional basis. This is clearly an approximation but if the two basis
functions are carrefully chosen, then the eigenfunctions obtained by diagonaliz­
ing the hamiltonian matrix in this 2-dimensional basis can be very accurate in
describing the ”exact” molecular orbitals. Let’s define the matrix elements of
the hamiltonian as:
�
� �
ˆ 1s (1)
E0 = φ1s (1)|Ĥ|φ1s (1) = φ∗1s (1)Hφ
which by symmetry is equal to:
�
� �
ˆ 1s (2)
E0 = φ1s (2)|Ĥ|φ1s (2) = φ∗1s (2)Hφ
and
V =
�
ˆ 1s (2) =
φ∗1s (1)Hφ
�
ˆ 1s (1)
φ∗1s (2)Hφ
The off-diagonal elements shown above are in general complex conjuguate to
each other, because the hamiltonian matrix must be hermitian, i.e Ĥ † = Ĥ.
Using E0 and V we find that the hamiltonian matrix expressed in the twodimensional basis of the φ1s (1) and φ1s (2) atomic orbitals is:
�
�
E0 V
ˆ
H=
V E0
The
φ1s (1) and φ1s (2) are real and normalized wavefunctions
� atomic orbitals
�
-i.e |φ1s (1)|2 = �φ1s (1)2 = 1 and the same for φ1s (2)- but they are not
orthogonal because φ1s (1)φ1s (2) = S =
� 0. In order to find the eigenergies
we will have to equate the determinant of the following matrix to zero (This is
a generalization of the determinantal eigenvalue equation for non-orthogonal
basis):
�
�
E0 − λ V − λS
V − λS E0 − λ
1. Find the eigenenergies (λ+ and λ− ) of an electron in an H2 molecule by
solving the determinantal equation above for λ. λ+ will be the smallest
eigenenergy (knowing that S is positive and V is negative) and λ− the
largest.
2
2. For each of the two possible eigenenergies λ+ and λ− , find the corre­
sponding normalized eigenvectors by solving the Schrodinger equation in
matrix form (This is a generalization of the matrix eigenvalue equation
for non-orthogonal basis):
�
��
�
E0 − λ ± V − λ ± S
c1
=0
V − λ± S E0 − λ±
c2
3. From the calculations carried out in the previous questions, we see that
an electron has two possible states in the H2 molecule. Because of the
spin degeneracy, the ground state of the H2 molecule consists of the two
electrons accupying the lowest energy state with opposite spins. Then
what is the total energy of the H2 molecule? (in its ground state)
4. If ĥ is the hamiltonian of an electron in the hydrogen atom, then φ1s is
an eigenfunction (the ground state) of ĥ for the eigenvalue �0 (the ground
state energy). Then what is the total energy of two isolated hydrogen
atoms? (in their ground state)
5. If one considers that �0 is equal to E0 (this is almost true for the H2
molecule), can you demonstrate that the formation of a H2 molecule is
favored energetically with respect to two isolated atoms?
3
The spin of the electron
During the 1920’s, physicists realized that in order for them to be able to explain
all the experimental results accumulated over the previous 20 years, they needed
to give to the electron an intrinsic property, the spin. In Quantum mechanics,
the spin of the electron is described by a 2 dimensional complex vector like:
�
�
a
|ψ� =
b
where a and b are complex numbers. In this exercise, we will consider that the
electron spin is described by the following quantum state:
�
�
√1
2
− √12
|ψ� =
Since the spin is a measurable quantity there are hermitian operators associated
�ˆ , which is a
with it. We have seen in class the angular momentum operator L
vectorial operator, i.e it has 3 coordinates (L̂x , L̂y and L̂z ) that are actually
operators of their own. Now suppose you want to know what is the y projection
of the spin of an electron described by the spin vector above (|ψ�). For this you
need the expression of the y projection of the spin operator Ŝy :
�
�
0 i
ˆ
Sy =
−i 0
Using the principles of Quantum mechanics, answer the following questions:
1. What are the possible measurement results for the y projection of the
spin?
3
2. Is |ψ� normalized?
3. Is |ψ� an eigenvector of the spin operator Ŝy ?
4. What are the eigenvectors of Ŝy ?
5. Can you express |ψ� in terms of the eigenvectors of Ŝy ?
6. What would be the probability of obtaining each one of the possible mea­
surement results for the y projection of the spin?
7. What is the quantum state of the particle after each possible measure­
ment?
4